I was in ADIPEC 2018 this week, where I meet several friends with whom I have worked in General Electric and as consultant in one of my latest projects.
A common theme that came across in our chats was the speed of change we are experiencing in our daily job.
The “horizon of our Present” is running faster than “our imagination of the Future”.
Whatever we think and do to cope with the last challenge at hand, we soon find that someone, somewhere, is far ahead of the curve in applying that very solution to quite similar tasks.
Our organization should be Fast and Resourceful, but often we find that these two requirements are a trade-off.
The strange thing is that in the not so far past, large organizations found a way to sort out this dilemma.
Faced with the fact that the number of input variable to a problem were increasing, CEO have started to recognize that their gut alone was not sufficient to decide, and more people should be brought on the decision table.
This has worked for a certain period of time, until the complexity of the decision to be taken has made explode the number of people sitting at that very table.
In a situation where the contribution of each factor to the solution went below the minimum threshold needed to clearly identify its causal effect, the move to increase the contributors to decision making has generated the paralysis:
In a lack of evidence, no one can claim of affecting the results, and therefore do-nothing is the only option where meeting participants find consensus.
It’s this lack of evidences that drives large organizations towards maintaining the status quo, slows decision making, and even refrains from using those abundant resources that are indeed available.
What happened is that the number of factors to be used in critical business decisions has exploded
this means that the contribution of each factor to the decision is barely visible - if visible at all.
What happened that destroyed the effectiveness of this decision making approach?
Good News: there is.
It’s called Artificial Intelligence, and its adoption is what is making the difference between companies that can decide fast and leverage their resources, and those who don’t.
Artificial Intelligence it’s exactly this:
Being able to process a zillion number of signals, and at the same time,
leverage their almost invisible contribution ( informative content ) to take reliable decisions fast.
They simply cannot play the game.
Even if they find a niche problem for which they can bet a solution faster than the average bureaucratic large organization, they cannot compete against the Artificially Intelligent equipped one.
Their destiny is to be left alive to become an outsourced part of the AI-Corporation.
This is already happening in sectors of the economy where “zombie” suppliers are a reality
( e.g.: search for the “poultry meat industry in USA” …).
If your organization is:
Probably it’s because it’s lagging behind in adopting Artificial Intelligence to support its decision making.
Before to start a journey that brings on board of your company a team of Data Scientist however, it would be beneficial to be sure that you ( or your team members ) are fully conscious of what it takes to embrace this new level of Decision Making.
On my experience this cannot be achieved by simply reading some introductury book or being told about AI by a Blue Chip consulting or one of your Peers in the industry.
The fact that the AI/ML level of Decision Making is achieved through a medium ("the computer") that skips a big chunk of the "Gut-processing" into Decision Making, poses a challenge to the management that till few days ago was not there: Statistically Reasoning.
Our brain in-fact is not made to reasoning "statistically".
Khaneman demonstrated it with the famous "Linda the Bank Teller" case, of 1983.
Khaneman told a sample of persons that:
"Linda is 31 years old, single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and also participated in anti-nuclear demonstrations."
Then he asked which of the following two alternatives is more probable:
Rationally, statement 2 cannot be more likely than statement 1, but a fully 85 percent of respondents said that it was.
Statistic's can be "faked". You cannot take decisions involving statistics using your Gut, your Experience, your Logical Tollgates or Your Communication skills.
(and tapping into your peers does not solve the problem, 'cause they are experiencing your same limitations)
The advent of Artificial Intelligence has brought to an end the game of practitioners and millantators:
So the only way to embrace AI/ML at its full is to add a new layer to the managerial skills:
that are in the current arsenal of a mangaer:
Not a general statistics however.
If you are trained on Statistical Process Control (six sigma) you start from a solid base, but is not enough.
You should make your hands dirty and embrace "numerical statistics": Machine Learning and AI itself.
Experimenting with a bit of Coding to handle Process Data through basic clustering algorithms, decision tree factors, Logistic regression classification etc.
will give you (or your managers ) a totally different perspective on your business model.
This will add your "business acumen" to Artificial Intelligence, and bring your company in the AI era.
Hope this helps, and as usual: call me to talk about this
The Blue Canvas is the starting point of our "Roadmap to Market" methodology,
that is helping industrial SMBs, high-tech Startups and University Spin-offs
bring innovation to market since 2009.
Read how it works here:
Get our booklet "the Roadmap to Market"
and build a business model that sells.
you have 90 days to tell us you did not like it
and we will give your money back
(and you keep the booklet)